SymmetricDS in the wild (production)?

Here is a blog article about SymmetricDS. JumpMind is working on re-branding their website which will include case studies for SymmetricDS. It should be live by the first of the year.

I ended up choosing SymmetricDS for my company after having investigated database replication/synchronization for quite some time. My main issue was that i have a database table which represents physical files that needs to be replicated along with the database. SymmetricDS even takes care of this.

The software is highly configurable, and the fact that you can plug in extension points (to take care of, for instance, file replication) makes it a powerful tool.

I haven't had it up and running very long, but it appears to be working very well. Even though the learning curve is steep in the beginning (with regard to configuration), i'd say that it is worth the time it takes to learn it!

UPDATE : I've had SymmetricDS running for almost 4 years now, and it has not failed a single time. It seems to be the Toyota Hilux of replication software : it just works. I'm replicating a local internal database to a remote external database. We're not talking about a huge amount of data to be replicated, but still there's a fair amount of activity going on. Replication is fast (seconds). Also, if for some reason you'd like to "populate" the external node from scratch, you can easily do this with SymmetricDS as well. Thus, you can use SymmetricDS from start and have it to first populate the external node from scratch as well as take care of continuous replication.

I've got replication set up both ways (for different tables, never tried same table), and this works very well as well.

UPDATE 2 : 8 years later and it's still running stable. I have not made much changes to the setup it is replicating, but it is certainly doing its job

We are using SymmetricDS between a MySQL server and 2 PostgreSQL servers (one of which is on a poor [<1Mbit] ADSL connection) fairly successfully on a small-ish database (pg_dump was 1.2G last week, ~15k rows INSERT/UPDATE'ed a day) .

Configuration can be hard to wrap your head around and we have/have-had some foreign key issues (which may not be due to the way we apply schema updates for this application).

Personally I found the few issues we had non-trivial and somewhat frustrating to debug, but we have had significantly fewer issues since upgrading to the 3.x releases (of which the remainder could be attributed to operator error rather than faults in the application).

My main missing feature is sync of the PostgreSQL sequences, but we work around that by setting the increment to 10 on all servers and offsetting each server to a different value (the MySQL server is primarily read-only, so ymmv there)